Subscriber access provided by Kaohsiung Medical University
A hybrid approach for selective sulfoxidation via bioelectrochemically derived hydrogen peroxide over a niobium(V)-silica catalyst James S. Griffin, Eric Taw, Abha Gosavi, Nicholas E Thornburg, Ihsan Pramanda, Hyung-Sool Lee, Kimberly A Gray, Justin M Notestein, and George F. Wells ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.7b04641 • Publication Date (Web): 27 Apr 2018 Downloaded from http://pubs.acs.org on April 27, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
ACS Sustainable Chemistry & Engineering
A hybrid approach for selective sulfoxidation via bioelectrochemically derived hydrogen peroxide over a niobium(V)-silica catalyst James Griffin1, Eric Taw1, Abha Gosavi1, Nicholas E. Thornburg1†, Ihsan Pramanda2, HyungSool Lee3, Kimberly A Gray4, Justin M. Notestein1, George Wells4* 1
Department of Chemical and Biological Engineering, Northwestern University, 2145 Sheridan Rd. E-136, Evanston, IL 60208. 2
Master of Science in Biotechnology, Northwestern University, 2145 Sheridan Rd. E-136, Evanston, IL 60208. 3
Department of Civil and Environmental Engineering, University of Waterloo, 200 University Avenue West, Waterloo, ON Canada N2L 3G1.
4
Department of Civil and Environmental Engineering, Northwestern University, 2145 Sheridan Rd. A-318, Evanston, IL 60208.
Keywords: microbial peroxide producing cell, bioelectrochemical system, green chemistry, catalysis, sulfoxidation, resource recovery
25 26
*Corresponding author: George F. Wells, A318 Technological Institute 2145 Sheridan Road
27
Evanston, IL 60208-3109, Phone: (847) 491-9902, Fax: (847) 491-4011, Email:
28
[email protected] 29
Submission as a Research Article to ACS Sustainable Chemistry and Engineering
30 31
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
32
Abstract:
33
In this work, we demonstrate a combined bioelectrochemical and inorganic catalytic
34
system for resource recovery from wastewater. We designed a microbial peroxide producing cell
35
(MPPC) for hydrogen peroxide (H2O2) production and used this bioelectrochemically-derived
36
H2O2 as a green oxidant for sulfoxidation, an industrial reaction used for chemical synthesis and
37
oxidative desulfurization of transportation fuels. We operated an MPPC equipped with a gas
38
diffusion electrode cathode for six months, achieving a peak current density above 1.4 mA cm-2
39
with 60% average acetate removal and 61% average anodic coulombic efficiency. We evaluated
40
several cathode buffers under batch and continuous flow conditions for solubility and pH
41
compatibility with downstream catalytic systems. During 24-hour batch tests, a phosphate
42
buffered MPPC achieved a maximum H2O2 concentration of 4.6 g L-1 and a citric acid-phosphate
43
buffered MPPC obtained a moderate H2O2 concentration (3.1 g L-1) at a low energy input (1.6
44
Wh g-1 H2O2) and pH (10). The MPPC-derived H2O2 was used directly as an oxidant for the
45
catalytic sulfoxidation of 4-hydroxythioanisole over a solid niobium(V)-silica catalyst. We
46
achieved 82% conversion of 50 mM 4-hydroxythioanisole to 4-(methylsulfinyl)-phenol with
47
99% selectivity with a 0.5 mol% catalyst loading in 100 minutes in aqueous media. Our results
48
demonstrate a new and versatile approach for valorization of wastewater through continuous
49
production of H2O2 and its subsequent use as a selective green oxidant in aqueous conditions for
50
green chemistry applications.
51
ACS Paragon Plus Environment
Page 2 of 30
Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Sustainable Chemistry & Engineering
52
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
53
Introduction:
54
Modern wastewater treatment is vital for protecting human and environmental health.
55
However, most design decisions are based on a cost benefit analysis that does not fully value
56
resource recovery. This has resulted in an unsustainable “once-through” model that focuses on
57
waste removal and disposal. Bioelectrochemical systems (BESs) use a biological catalyst to
58
produce electrical current from wastewater.1 Most BESs have focused on producing electricity in
59
microbial fuel cell (MFC) configurations; however, there are some examples of cathodic
60
electrochemical synthesis including hydrogen (H2) or hydrogen peroxide (H2O2) production in
61
microbial electrolysis cells (MECs) or microbial peroxide producing cells (MPPCs),
62
respectively.2-3 A life cycle assessment of MPPCs found that at scale, MPPC-derived H2O2 has
63
nearly 60% lower life cycle greenhouse gas emissions than H2O2 produced via the traditional
64
anthraquinone auto-oxidation process.4 Previous studies on H2O2 production in MPPCs5-7 have
65
focused on producing alkaline H2O2 for disinfection, industrial bleaching or non-specific
66
advanced oxidative processes such as the bio-electro-Fenton process for removal of recalcitrant
67
contaminants from water8.
68
In addition to its proven uses in disinfection and bleaching, H2O2 is a promising green
69
oxidant for sustainable chemistry applications because, unlike organic hydroperoxides, it
70
generates water as its only byproduct.9-11 While the use of H2O2 with solid oxide catalysts to
71
selectively oxidize sulfides has been extensively studied,12-15 there are few examples of selective
72
oxidation in these systems in aqueous conditions, and none involving bioelectrochemically
73
derived peroxide.16 We recently demonstrated high rates and selectivities in the oxidation of
74
thioanisole and several benzothiophene derivatives by H2O2 over a highly dispersed supported
75
niobium(V)-silica (Nb(V)-SiO2) catalyst in acetonitrile, which outperformed benchmark titania-
76
silica and zirconia-silica materials.17 Thus, sulfoxidation of a thioanisole derivative over a
77
similarly formulated niobium(V)-silica catalyst was chosen as a model reaction system for proof-
78
of-concept testing of the direct use of bioelectrochemically derived H2O2 for selective oxidation
79
in an aqueous buffer (rather than in acetonitrile or other organic solvents). This approach could
80
be adapted to the epoxidation and/or dihydroxylation of alkenes,18 the oxidative
81
depolymerization of lignin macromolecules19, or other H2O2-dependent catalytic processes.
82
H2O2 can be generated for these applications via electrocatalytic oxygen reduction. In
83
BESs, anode respiring bacteria anaerobically consume organic matter using an external electrode
ACS Paragon Plus Environment
Page 4 of 30
Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Sustainable Chemistry & Engineering
84
as a terminal electron acceptor via a metabolic process known as extracellular electron transfer
85
(EET).20 Current produced during EET can reduce oxygen to water via a four-electron reduction
86
(Equation 1). Oxygen can also be partially reduced to H2O2 via a two-electron oxygen reduction
87
reaction that occurs on graphitic cathodes (Equation 2).21 H2O2 can be reduced further, resulting
88
in a net production of water and decreased coulombic efficiency in an MPPC.22
89
+ 4 + 4 → 2 = 0.81V23
(Eq. 1)
90
+ 2 + 2 → = 0.28V23
(Eq. 2)
91
H2O2 synthesis in acetate-fed MPPCs is exergonic,7 although most previous efforts have applied
92
external voltage to improve H2O2 synthesis rates and titers at the cost of increasing energy
93
intensity. Rozendal et al. developed the first bioelectrochemical H2O2 reactor and achieved
94
concentrations of 1.9 g L-1 at an efficiency of 83% and energy input of 0.93 Wh g-1 H2O2 using
95
an acetate media as feed.5 Fu et al. demonstrated that H2O2 could be generated with a net positive
96
energy production but achieved a maximum concentration of 79 mg L-1.7
97
System design improvements such as minimizing electrode spacing3 and the use of
98
composite carbon black-graphite-PTFE cathodes24-26 have led to higher efficiency and lower
99
system overpotentials. Despite recent improvements in system performance and cell design, “pH
100
splitting”, or opposing pH shifts at the anode and cathode chambers, remains a major problem
101
limiting MPPC H2O2 titer. Proton consumption during oxygen reduction raises catholyte pH and
102
commonly used ion exchange membranes primarily transport ions other than hydroxide or
103
protons at typical ionic strengths.27 Elevated cathode pH decreases coulombic efficiency by
104
promoting H2O2 decomposition28 and increases cell overpotential in MPPCs, thus increasing
105
energy input per gram of H2O2 produced. In a recent study, 80% of the initial H2O2 stored in pH
106
12.5 NaCl electrolyte solution decomposed in three days.3 For the latter reason, there is
107
significant motivation for the immediate use of the produced H2O2 in a continuous process.
108
In this study, we present a novel hybrid biological and chemical catalytic approach to
109
wastewater resource recovery via bioelectrochemical H2O2 production and subsequent use of the
110
H2O2 solution as a green oxidant in chemical synthesis via heterogeneous catalysis. Specifically,
111
we demonstrate thioether sulfoxidation, an industrially relevant process for the synthesis of
112
medicinally relevant sulfoxides and sulfones29 and for the removal of sulfur compounds from
113
fuels and industrial effluents. First, we optimized MPPC cell design, hydraulic residence time
114
and cathode buffer composition to maximize H2O2 concentration at a pH compatible with a
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
115
silica-based catalyst. We operated an MPPC using a continuous flow cathode for six months and
116
produced H2O2 at average effluent concentrations of 3.2 g/L at a 40-hour HRT. Next we
117
characterized the kinetics and selectivity of catalytic sulfoxidation of 4-hydroxythioanisole to 4-
118
(methylsulfinyl)-phenol by H2O2 in aqueous buffer solutions and used that to design and operate
119
a system for continuous sulfoxidation using the MPPC effluent.
120 121
Methods:
122 123
MPPC Reactor Configuration:
124
Bioelectrochemical cells were constructed from laser-cut acrylic with an anode chamber
125
volume of 200 mL equipped with an Ag/AgCl reference electrode (BASI, West Lafayette, IN).
126
During MPPC operation, a 25 cm2 piece of AvCarb carbon felt (Fuel Cell Earth, Woburn, MA)
127
with a copper mesh current collector was used as the anode. The carbon felt was pretreated for
128
24 hours in 1N nitric acid, followed by 24 hours in acetone and 24 hours in ethanol.30 During
129
abiotic cathode optimization, the copper-backed carbon felt was replaced by a 10 cm2 Pt mesh
130
(Sigma) that was used as a counter anode. The cathode chamber had a 100 mL chamber with an
131
exposed cathode surface area of 25 cm2. A gas diffusion cathode was constructed as previously
132
described24 using a carbon cloth electrode (GDL-CT, Fuel Cells Etc., College Station, TX)
133
coated with PTFE on one side and 5 mg cm-2 carbon black (Vulcan XC-72, Fuel Cell Earth,
134
Woburn, MA) dispersed in Nafion on the other side. The chambers were separated by a 33 cm2
135
anion exchange membrane (Ultrex AMI-7001, Membranes Intl., Ringwood, NJ). The electrode
136
spacing was 3.4 cm. Both chambers contained inlet and outlet ports to operate in continuous
137
mode. During continuous H2O2 production and usage, cathode effluent was stored in a holding
138
tank and neutralized and diluted before being used for sulfoxidation. A schematic of the
139
combined MPPC and sulfoxidation reactor setup is shown in Figure 1, and photographs of the
140
reactor setups are available in Figure S1.
141 142
MPPC Operating Conditions
143
The anodic biofilms used in this study were inoculated with 80 vol% synthetic acetate
144
media, 10 vol% primary effluent from the O’Brien Water Reclamation Plant (Skokie, IL), and 10
145
vol% effluent from a previously operated MPPC. The anode potential was set at -0.3V vs.
ACS Paragon Plus Environment
Page 6 of 30
Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Sustainable Chemistry & Engineering
146
Ag/AgCl using a VMP-3 potentiostat (Bio-logic USA, Knoxville, TN) and cell current, working
147
and counter electrode potentials were monitored every 30 seconds. The anode was fed with a
148
synthetic wastewater medium containing 25 mM CH3COONa in a 100 mM phosphate buffer31.
149
Each liter of anode medium also contained 0.3 mL of a trace metal solution.32 MPPC anodic
150
hydraulic residence time (HRT) was initially 12 hours and then decreased to 5 hours as acetate
151
consumption rates increased to avoid substrate limitation.33 The medium was continuously
152
sparged with N2 gas to minimize oxygen introduction into the system. The reactor was operated
153
continuously for over six months, during which time cathode HRT and buffer concentration were
154
varied to study the effect of these variables on efficiency and effluent H2O2 concentration. All
155
MPPC experiments were performed at room temperature (22 ±1°C) with continuous mixing.
156
To identify buffer systems compatible with downstream catalytic processes, H2O2
157
production was tested in abiotic cells equipped with Pt anodes. H2O2 production rates, cathodic
158
coulombic efficiency and catholyte pH increase were evaluated in 24 hour batch assays with
159
several phosphate and citric acid buffer compositions, listed in Table S1. Cathode potential was
160
fixed at -0.6V vs. Ag/AgCl and samples were collected for pH and H2O2 measurement. During
161
continuous MPPC operation, catholyte chambers were triple rinsed with deionized water and
162
then allowed to accumulate H2O2 for 24 hours before starting continuous flow. Effluent H2O2
163
concentrations were monitored until they reached steady state, and effluent was collected for
164
sulfoxidation.
165 166
Analytical Methods:
167
A Trace 1310 gas chromatograph (GC) (Thermo Scientific, USA) equipped with a
168
diphenyl dimethyl polysiloxane column and an FID detector was used to measure anolyte acetate
169
concentrations. A 0.1 µL sample was injected in the GC with a 50:1 split ratio under a constant
170
flow rate of 1.0 ml min-1. The GC oven was held at 70 °C for one minute, then increased at 10 °C
171
per minute up to 180 °C, and then held at a constant temperature of 180°C for five minutes. 24
172
hour anodic acetate removal rates were calculated by measuring the difference between acetate
173
medium and effluent concentrations at a given HRT. H2O2 concentration was measured in the
174
catholyte effluent using the ammonium metavanadate method.34 A vanadate colorimetric reagent
175
was prepared by slowly adding 9 M sulfuric acid to a 12.4 mM ammonium metavanadate
176
solution under magnetic stirring at 50°C until complete dissolution. Triplicate 0.75 mL catholyte
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 30
177
samples were diluted to 20-200 mg L-1 H2O2 in deionized water and added to an equal volume of
178
the vanadate solution. Absorbance was measured at 450 nm on a BioSpectrometer UV-Vis
179
spectrophotometer (Eppendorf). Anodic and cathodic coulombic efficiencies were calculated
180
based on the total current and measured changes in acetate and H2O2 concentrations using
181
equations 3 and 4.23
182
Anode coulombic efficiency (%):
183
Cathode coulombic efficiency (%):
184
where I is the total current, n is the electron equivalents per mole of H2O2 or acetate (2 and 8,
185
respectively), [CH3COO-] and [H2O2] are the concentrations of acetate and H2O2, respectively, in
186
mol L-1, V is the anode volume and Q is the volumetric flow rate of catholyte. Overall coulombic
187
efficiency was calculated by multiplying the anodic and cathodic coulombic efficiencies
188
together. Aerial current and power density were calculated by normalizing current and power to
189
the anode and cathode surface area (25 cm2). Batch cathodic coulombic calculations were
190
performed using the volume of catholyte remaining in the reactor while sampling to account for
191
changes in catholyte volume during the batch assays.
∆[ ]"#$%&' [( ( ])
(Eq. 3) (Eq. 4)
192 193
Catalyst Synthesis and Characterization:
194
The Nb–SiO2 catalyst was prepared by grafting a niobium(V)–calixarene coordination
195
complex to partially-dehydroxylated silica gel (Alfa Aesar, 75-150 µm, 373 m2 g-1 BET)
196
following similar procedures recently reported by some of us and described in the Supporting
197
Methods.35-36 Catalyst metal content was quantified using inductively coupled plasma atomic
198
emission spectroscopy (ICP-AES; Thermo iCAP 7600). Prepared Nb(V)-SiO2 catalysts were
199
digested in 48 wt % HF, diluted with 0.9 wt % HNO3 and calibrated against a standard curve of a
200
commercial Nb solution (Fluka Analytical) in 0.9 wt % HNO3. Diffuse reflectance UV-visible
201
(DRUV-vis) spectra were collected from 800–200 nm at ambient conditions using a Shimadzu
202
UV-3600
203
polytetrafluoroethylene (PTFE) powder as a baseline perfect reflector and a 20:1 catalyst diluent.
with
a
Harrick
Praying
Mantis
diffuse
reflectance
accessory,
using
204 205 206
4-hydroxythioanisole Oxidation: 4-hydroxythioanisole oxidation kinetics were first investigated across a range of
ACS Paragon Plus Environment
Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Sustainable Chemistry & Engineering
207
temperatures, catalyst loadings and aqueous buffer conditions using commercial H2O2 in a
208
“mock MPPC catholyte” as follows: 1 mmol 4-hydroxythioanisole (Sigma) and 0.1 mmol phenol
209
(internal standard, Sigma) were dissolved in a 2:1 deionized water: ethanol solution. Mock
210
MPPC catholyte was produced by pH adjusting a 200 mM pH 6 phosphate buffer (21 g L-1
211
KH2PO4 (Sigma) and 6.6 g L-1 Na2HPO4 . (H2O)7 (Sigma)) to pH 13 with NaOH and then
212
neutralized with HCl to pH 7 to mimic the neutralization procedure used for the MPPC catholyte
213
and achieve a similar chloride concentration in the mock and actual MPPC catholyte. 31.1 mg
214
Nb(V)-SiO2 (0.16 mmol Nb g-1) was then added to a solution containing 15 mL of the 4-
215
hydroxythioanisole solution and 5 mL of mock MPPC effluent. Finally, 110 µL of H2O2 (30 wt
216
% aqueous, 1.07 mmol) was added to start the reaction. Experiments were run between 25°C and
217
75°C on a hotplate and magnetically stirred at 500 rpm. At the start of reaction, the 4-
218
hydroxythioanisole concentration was 50 mM and the molar ratios inside of the reactor were
219
100:107:0.5 for 4-hydroxythioanisole:H2O2:Nb. Reaction aliquots (approx. 200 µL) were
220
quenched with approximately 5 mg NaHSO3 (Sigma) at desired time points before GC analysis.
221
Product identities were initially determined using GC-MS (Shimadzu QP2010, Zebron ZB-624
222
capillary column) and quantified using GC-FID (Shimadzu 2010, TR-1 capillary column) via
223
calibrated standards. The reaction was assumed to be first order in both H2O2 and 4-
224
hydroxythioanisole, giving second order overall, and rate constants were found by a satisfactory
225
linear regression of ln
226 89
,-. ,/0-. 1
= 23 /4 − 1167
(Eq. 5)
227
where 4 =
228
H2O2, respectively. Apparent activation barriers were calculated from the slope of ln(k) vs.
229
1000/T (K) conforming to an Arrhenius model.
.9
= 1.1, k is the apparent rate constant, and A and B as 4-hydroxythioanisole and
230
4-hydroxythioanisole oxidation kinetics were quantified with MPPC-derived H2O2 using
231
a similar procedure. The reactor sampling and GC quantification were as described above, but
232
the mock MPPC catholyte and 30 wt% H2O2 were replaced with H2O2-laden MPPC effluent
233
adjusted to pH 7 with HCl. The volumes of MPPC effluent and deionized water added to the 4-
234
hydroxythioanisole solution were adjusted to give initial molar ratios of 100:110:0.5 for 4-
235
hydroxythioanisole:H2O2:Nb in the reactor.
236
In addition to batch experiments, continuous stirred tank reactor (CSTR) sulfoxidation
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 30
237
experiments were performed in a 50 mL stirred round bottom flask controlled at 35 ºC with a 20
238
mL holdup volume. Separate MPPC effluent and 4-hydroxythioanisole solutions were pumped
239
into the reactor at 5 mL hr-1 each using a Minipuls 3 multichannel pump (Gilson, Inc., Middleton
240
WI) to give a two-hour HRT. H2O2 was generated in an MPPC operated at sufficiently long
241
HRTs (on average 1.7 days) in order to reach 110 mM H2O2 (3.7 g H2O2 L-1). MPPC effluent
242
was acidified with HCl to pH 7 prior to usage. The sulfide reactant solution initially contained
243
100 mM 4-hydroxythioanisole and 40 mM phenol dissolved in a 50:50 solution of water and
244
ethanol. The reactor was initially filled with 10 mLs of 4-hydroxythioanisole and catholyte
245
solutions each and contained 0.5 mol% Nb(V)-SiO2 (relative to 4-hydroxythioanisole). Cotton
246
dead-end filters were used to minimize catalyst loss through the effluent ports on the CSTR.
247
Effluent sulfide product distribution (unreacted sulfide [4-hydroxythioanisole], product sulfoxide
248
[4-(methylsulfinyl)-phenol], and sulfone [4-(methylsulfonyl)-phenol] concentrations) was
249
measured as in the batch experiments, and the reactor was operated until the product distribution
250
reached steady state. The apparent rate constant was determined from the effluent concentration
251
from the CSTR using the Levenberg-Marquardt nonlinear least squares regression algorithm and
252
the Runge-Kutta method for numerically integrating differential equations in MATLAB R2016b
253
(The Mathworks, Natick MA). The nlparci function was used to obtain a 95% confidence
254
interval. The effluent concentration was fit to the following differential equation, derived from a
255
mass balance:
256 257
.
=
.9 . :
− 623 /2; − 23 + 23 1
(Eq. 6)
where A and B are 4-hydroxythioanisole and H2O2, respectively.
258 259
Results and Discussion
260 261
Microbial Peroxide Producing Cell (MPPC) operation and H2O2 production
262
We operated two MPPCs with a fixed anode potential of -0.3V (vs. Ag/AgCl) and gas
263
diffusion electrodes optimized for H2O2 production for six months. Bioelectrochemical current
264
began after 5 days and rapidly entered an exponential growth phase. Current density peaked 10
265
days after inoculation at 1.4 mA cm-2 but declined to an average of 0.8 ± 0.3 mA cm-2 for the
266
duration of the experiment. Long-term cell current for one cell is shown in Figure S2. Consistent
267
with previously reported long-term BESs, differences in current maxima before and after
ACS Paragon Plus Environment
Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Sustainable Chemistry & Engineering
268
replacing media were apparent after short downtime periods.37 Current density varied during
269
operation due to recurring batch acetate removal experiments, feed interruptions, and
270
maintenance breaks during operation. After startup, average acetate removal was 8.7 ± 2.7 mg
271
cm-2 day-1 (normalized to anode surface area), with a anodic coulombic efficiency of 61 ± 22%
272
(Figure S3). During operation, attached growth on the sides of the reactor was observed, likely
273
due to the growth of microaerobic bacteria that could grow under the hypoxic reactor conditions
274
(